Current therapeutic regimens have failed to decrease the high (80 to 90%) mortality rate of neuroblastoma, indicating the need for new therapeutic approaches. Crucial to any approach to therapy is the knowledge of the cell biology of neuroblastoma. One important characteristic of this tumor, both in the patient and in tissue culture, is its marked cellular heterogeneity. Previous and current studies show that many neuroblastoma cell lines comprise two or more cell types: neuroblastic (N) cells which contain neurofilaments and noradrenergic properties (enzymes, uptake mechanisms, and receptors) and substrate-adherent (S), non-neuronal cells with enzymes, intermediate filaments, cell surface antigens, and extracellular matrix (ECM) proteins characteristiC Of embryonic mesectodermal (glial/melanocytic/meningeal) cells. Recently identified intermediate (I) cells share properties of both N and S cells and may represent either a stem cell or a transitional cell between N and S. Important findings related to this cellular heterogeneity are: (i) N and S cells undergo bidirectional phenotypic interconversion (transdifferentiation); (ii) morphological, biochemical, and cell surface antigen changes in N/S interconversion are tightly coordinated, suggesting that this process is controlled by one or very few regulatory genes; and (iii) N cells are highly tumorigenic whereas S cells are non-tumorigenic. Two other pediatric tumors, Ewing's sarcoma and peripheral neuroepithelioma, appear neuroectodermal in origin and comprise N- and S-like cells. Proposed aims are to: (1) quantitate expression (and gene copy number) in N and S cells of oncogenes N-myc and Ha-ras, EGF and NGF receptor and TGF alpha, cell surface antigens, and gangliosides by molecular hybridization and/or immunochemical techniques as a means of distinguishing correlates of both differentiation state and transformation state; (2) examine effects of several soluble-, growth and differentiation factors (e.g., retinoic acid and cyclic AMP) as a means of experimentally inducing differentiation; (3) study t-role of the ECM and cell-cell interactions in N/S interconversion, in an attempt to identify extrinsic signal(s) regulating differentiation or transdifferentiation in clonal cell lines; and (4) determine whether tumors in vivo contain counterparts of the S observed in cell culture to evaluate the prevalence of N/S transdifferentiation in cal neuroblastoma. The current aims represent a continuing effort to define mechanisms underlying transdifferentiation and transformation in human neuroblastoma.